Marco Conceptual

Abstract

The increase in the production of distiller’s dried grain with soluble (DDGS), a major co- product of corn ethanol process, provides an excellent opportunity to develop value-added products. The fibers from DDGS, separated by elusieve process, were treated in a sodium hydroxide (NaOH) solution to investigate their performance as fillers in high density polyethylene (HDPE) composites. The composite material properties were characterized by water absorption, tensile, flexural, and impact tests. Composite samples were manufactured with NaOH treated DDGS fiber at 25% and 50% loading. Addition of both untreated and NaOH treated 25% DDGS fiber to HDPE resulted in moisture absorption of less than 5%. At 50% fiber loading, composites with untreated DDGS fiber exhibited 15% moisture absorption compared to 25% moisture absorption observed for alkali treated DDGS fiber, after 30 days of water

exposure. The addition of treated and untreated DDGS fibers increased the tensile modulus but decreased the tensile strength of composites compared to neat HDPE. Under flexural load, the composite with 25% alkali treated DDGS fiber composite samples showed higher flexural modulus and flexural strength than untreated 25% DDGS composites. An increase in fiber loading from 25% to 50% decreased impact resistance of composites with both untreated and alkali treated DDGS fibers. The thermal stability of composites with alkali treated DDGS fibers increased in the temperature range of 150-230 °C, corresponding to hemicellulose degradation, in comparison to those with untreated DDGS fiber.

Introduction

The depletion of petroleum resources and negative impacts of conventional synthetic fiber composites on environment have stimulated the interest of automobile, aircraft, building and packaging industries to replace petroleum based analogs with sustainable materials. Natural fibers are a suitable replacement for synthetic fibers in polymer composites. Natural fibers have several advantages such as abundance, low cost, biodegradability, high sound absorption, fracture resistance, low density, acceptable specific strength and easy processing [1, 2].

The natural fibers are polar in nature, whereas polymers are non-polar materials. This mismatch in properties results in poor compatibility between the fiber and the polymer matrix, resulting in poor interfacial bonding and high moisture absorption by fibers. Weak interfacial adhesion prevents the stress transfer from the matrix to the fiber under applied load. The

interaction between the fibers and matrix can be improved by changing in the surface properties of the fiber. Natural fibers can be modified through physical or chemical treatments. Physical treatments change structural and surface properties of the natural fibers by changing the surface energy, thus increasing their compatibility with the polymer and improving their mechanical bonding to polymers. Different physical treatments such as corona discharge [3], cold plasma [4], gamma-ray [5] and UV bombardment [6] and chemical treatments such as mercerization (alkali) [7-9], grafting [10-12], acrylation [13], permanganate [14], acetylation [15-17], silane [18] and peroxide [19] have already been used effectively on the natural fibers.

Alkali treatment is a simple, inexpensive and effective method for surface modification of fibers. Alkali treated fibers have been reported to show increased surface area, which in turn leads to better interfacial bonding with polymer matrix [8,9].

Distillers’ dried grain with solubles (DDGS) is the main co-product produced from corn ethanol fermentation. The production of DDGS has seen tremendous increase in the last few

years [20]. The fiber fraction of DDGS isolated by a physical separation process named elusieve [21]. This separated DDGS fibers have the potential to be used as filler in polyolefin composites [22].

Currently, there are no reported studies that investigated how mercerization or sodium hydroxide (NaOH) treatment of DDGS fiber affects their performance as fillers in HDPE composites. The aim of this study was to investigate the performance of NaOH treated DDGS fibers compared to untreated DDGS fibers as fillers in HDPE composites.

Materials and Methods Materials

The DDGS was supplied by Midwest Ag Group (Underwood, ND), and the DDGS fibers were separated using an Elusieve process at Mississippi State University (Starkville, MS). The polymer used was Marlex 9006 high density polyethylene (HDPE), manufactured by Chevron Philips (The Woodlands, TX). The neat HDPE had a density of 0.952 (g/cm3), melt Index of 5.2 g/10 min measured at 190 °C with 2.16 kg, tensile yield strength of 18 MPa and flexural

modulus of 612 MPa. The sodium hydroxide (NaOH) pellets were obtained from Sigma Aldrich (Fargo, ND).

Mercerization

The DDGS fiber was first sized and screened through a 1 mm sieve using a Wiley mill (Model 4, Thomas Scientific, Swedesboro, NJ). The material was again screened to obtain fiber in the range of 0.250 mm-0.595 mm (30-60 mesh) using a Ro-Tap shaker (W.S. Tyler® Ro- Tap® 8in Sieve Shaker, Mentor, OH). After screening, the DDGS fiber was dried in an oven for 24 h at 105 °C to bring the moisture content below 1%.

The DDGS fibers were mercerized by immersing the fibers in a 0.1M NaOH solution in a 500 ml Erlenmeyer flask, kept at room temperature (~24 °C) for 1 h. After the 1 h treatment, the

fibers were washed by the tap water until the pH observed was 7, indicating all the NaOH residue was removed. The treated fibers were next oven dried at 105 °C for 24 h. The

composition analysis of untreated and NaOH treated DDGS fibers was performed by AOAC methods [23].

Manufacturing of Composites

A laboratory experiment was conducted with two fiber treatments (NaOH treated and untreated), and two fiber loading (25% and 50% by weight) factors, resulting in four types of composites. Samples were also made from neat HDPE for comparison. To manufacture

composite samples, the dried fibers were mixed with the HDPE at two different fiber loadings of 25 and 50 weight % and compounded in a twin-screw co-rotating extruder (L/D ratio of 18, Leistriz Micro 18 GL/-40 D, Allendale, NJ). The extruder barrel zones were set at temperatures between 160-195 °C and the die temperature was set at 195 °C. The extruder was operated at 150 rpm. The material was extruded into 3 mm diameter strands that were cooled through a water bath, before pelletizing with a BT25 pelletizer (Scheer Bay Co., Bay City, MI). The pelletized composite material was oven dried overnight at 80 °C before injection molding into test

specimens. The test specimens were manufactured with an injection molder (Model SIM- 5080, Technoplas Inc., Norwood, MA) set at 190 °C into dog bone samples of 12.5 mm by 3.5 mm cross sectional size at the center, and 65 mm length. The composite samples were stored in a sealed plastic bags before performing various tests.

Fiber Characterization

The morphologies of untreated and alkali treated DDGS fibers were examined under a Leitz Laborlux microscope (Laborlux S, Leitz, Wetzlar, Germany) for fiber surface

The thermal properties of the NaOH treated and untreated DDGS fibers were determined using a Q500 Thermal Gravimetric Analyzer (TA instruments, New Castle DE, USA). About 30 mg of fibers were placed in a platinum pan and heated from 25 °C to 800 °C at a ramp rate of 20 °C/min under a 60 mL/min air flow. The results were analyzed with TA instruments Universal Analysis software.

Characterization of Physical and Mechanical Properties of Composites

Melt Flow Index (MFI)

The MFI of the neat HDPE and the composite materials was determined in accordance with ASTM D1238 [24] standard using an extrusion plastometer (Tinius Olsen, Model MP 600, USA). The HDPE and composite pellets were tested in five replicates at 2.16 kg load and 190 °C. The MFI was recorded as the amount of material that would pass through the nozzle of the plastometer in 10 min.

Water Absorption

The long term water absorption of composites was quantified as specified by ASTM D570 [25] standard. The samples were immersed in water at 24 °C in a water bath. Five samples from each formulation were tested for percentage moisture gain at 24 h increments, and the test continued until the sample weight change stabilized. The composites used for the test were 30 mm long, 12.5 mm wide, and 3.5 mm thick.

Tensile Properties

The tensile properties such as stiffness and strength of the samples were measured according to ASTM D638 [26]. The samples used for the test were 63 mm long, 10 mm wide in the center and 3.5 mm thick. The crosshead speed was set at 5 mm/min. The universal testing machine Instron (Model 5567, Norwood, MA, USA) installed with a 2 kN load cell was used for the test.

Flexural Properties

The flexural properties such as stiffness and strength of the samples were measured according to the ASTM D790 [27] standard that specifies three-point bending test method for unreinforced and reinforced plastics. The samples used for the test were 63 mm long, 12.5 mm wide and 3.5 mm thick. A crosshead speed of 1.3 mm/min was selected based on the span length of the samples. The flexural properties were evaluated using universal testing machine as

described earlier.

Impact Strength

The impact strength of the samples was tested in accordance with ASTM D256 [28] standard. The notched samples were tested for their impact resistance properties using an Izod impact tester (Tinius Olsen, Model Impact 104, Horsham, PA, USA). The sample dimensions were 63 mm x 12.5 mm x 3.5 mm (LxWxH) with a notch of 2 mm. The microscopy images of the impact fractured sample were examined under a Leitz Laborlux S digital microscope at 40X magnification.

Statistical Analysis

To compare the properties of composites from different groups, Fishers Least Square Difference (LSD) tests were performed on all the 10 treatments using Minitab 17 (Minitab Inc., Penn State University, PA, USA). The error bars in the bar graphs represent the standard deviation of the group.

Results and Discussion

Composition Analysis of Untreated and NaOH Treated DDGS Fibers

The composition analysis of DDGS fibers showed that NaOH treated DDGS fibers had higher NDF concentration due to solubilization and loss of hemicelluloses and lignin during alkali treatment (Table 4.1). The alkali treatment of 0.1 M for a residence time of 1h proved to be

strong enough for solubilization of protein,starch, and fat contents that was present in the untreated DDGS fibers.

Table 4.1. Composition of untreated and NaOH treated DDGS fibers computed on a dry basis, performed by Animal Sciences lab (NDSU)

Morphology of Untreated and Alkali Treated DDGS Fibers

The NaOH treated fibers exhibited cleaner untreated DDGS fibers (Fig. 4.1a &b). The effects of alkali treatment include solubilization of hemicelluloses, lignin, wax, and oil covering the surface of the fiber [29]. The yellowish appearance of untreated DDGS fiber was mainly from the oil present on the surface which was not visible for the alkali treated DDGS fiber. The alkali treated DDGS fibers showed rough surface and more cellulose exposure which helps in improving the fiber-matrix adhesion.

Fig. 4.1. The microscopy images of a. untreated DDGS fiber and b. alkali treated DDGS fiber at 40X magnification

Components Untreated DDGS fiber (%) NaOH treated DDGS fiber (%)

NDF 53.20 76.8 Protein 34.21 15.62 Starch 4.04 1.54 Fat 4.04 0.81 Ash 4.45 3.02

a

_b

Thermogravimetric Analysis of DDGS Fibers

The thermogravimetric curves showed difference in the degradation rate of alkali treated DDGS as compared to the untreated DDGS fiber (Fig. 4.2 & 4.3). Alkali treated DDGS fiber lost 10% weight around 100 °C, representing moisture loss from the fiber. The thermal degradation of alkali treated DDGS fiber remained constant between 150 and 230 °C

temperature range, corresponding to hemicellulose degradation. The higher degradation in alkali treated DDGS fiber in the hemicellulose range is an indication that mercerization partially solubilized the hemicellulose in treated DDGS fiber. Between 230 and 360 °C temperature range, corresponding to cellulose degradation, the alkali treated and untreated DDGS fiber showed weight loss of approximately 64 and 68% respectively. At the end of the test, DDGS fibers showed a final weight loss percentage of around 75%.

Fig. 4.2. Thermogravimetric curves for untreated and alkali treated DDGS fibers

0 10 20 30 40 50 60 70 80 90 100 0 100 200 300 400 500 600 700 800 Weig ht (%) Temperature (°C) Alkali treated DDGS Untreated DDGS

Fig. 4.3. Differential thermogravimetric curves for untreated and alkali treated DDGS fibers showing their thermal degradation

Melt Flow Index (MFI) of DDGS composites

The addition of DDGS fiber in HDPE polymer decreased its MFI as expected (Table 4.2). The addition of 25% untreated or NaOH treated DDGS reduced the MFI from 5.2 to 3.93 and 3.5 g/10 min, respectively. The MFI decreased further to 1.7 and 0.98 g/10 min for untreated and NaOH treated fibers respectively when the loading of fibers was increased to 50%. The presence of fibers hampers the ability of HDPE chains to flow smoothly, thus reducing the MFI.

Table 4.2. Melt flow index of neat HDPE and untreated and NaOH treated DDGS composites

* + indicates the standard deviation of the treatment

** letters in parentheses indicate significant difference between the treatments at α = 0.05

-20 -15 -10 -5 0 0 100 200 300 400 500 600 700 800 Der iv a tiv e w eig ht (%/ °C) Temperature (°C) Untreated DDGS Alkali treated DDGS

Sample Formulations MFI- Untreated (g/10 min)

MFI- Alkali treated (g/10 min)

Neat HDPE 5.2 + 0.40 (a)*

DDGS25 (25% DDGS fiber) 3.93 + 0.20 (b)** 3.51 + 0.18 (c) DDGS50 (50% DDGS fiber) 1.7 + 0.16 (d) 0.98 + 0.31 (e)

Water Absorption of Composites

The composite samples with 25% untreated or alkali treated DDGS fiber absorbed less than 5% moisture after an exposure time of 30 days (Fig. 4.4). The low moisture absorption by the DDGS25 samples indicates that the HDPE polymer sufficiently encapsulated DDGS fibers, thus inhibited the entry of moisture in the composites. When the DDGS fiber loading was increased to 50%, the composite samples exhibited high moisture absorption after the initial 24 h. The untreated DDGS50 samples showed consistently higher moisture absorption than the alkali treated DDGS50 up to 264 h after which the alkali treated DDGS50 samples started to absorb moisture more rapidly than the untreated samples. The final moisture absorption of alkali treated DDGS50 samples was around 25% compared to 15% observed for the untreated

DDGS50 sample after 30 days. The alkali treatment removes lignin, pectin, waxy substance and natural oil covering the surface of cell wall, thus exposing cellulose fibrils [30]. Change in morphology may create more void to water penetrate inside the alkali treated samples. It is also possible that the fiber encapsulation was relatively poor for composite with 50% fiber loading h contributing to higher water absorption.

Fig. 4.4. The moisture absorption of untreated and alkali treated DDGS composite samples at 25% and 50% fiber loadings plotted against square root of time

Tensile Properties of DDGS Composites

The inclusion of DDGS fiber in HDPE matrix resulted in significant increase in the tensile modulus (Fig. 4.5). The DDGS25 composite samples exhibited increased stiffnesses that were 35% and 46% higher for the untreated and alkali treated samples respectively than the neat HDPE sample. This increase was most likely due to higher tensile stiffness of DDGS fiber and fiber encapsulation by matrix at 25% fiber loadings. When the fiber loading increased to 50%, the tensile stiffness of the untreated and NaOH treated DDGS50 samples showed an increase of 51% and 142%, respectively, compared to the neat HDPE. The alkali treated DDGS50 composite had the highest tensile stiffness of 1428MPa. In addition, alkali treated DDGS25 sample showed tensile stiffness comparable to the untreated DDGS50 sample.

0 5 10 15 20 25 30 0 5 10 15 20 25 30 M oistu re ab sor p tion (% ) Time (h1/2₎ Untreated DDGS 25% Alkali treated DDGS 25% Untreated DDGS50% Alkali treated DDGS 50%

Fig. 4.5. Tensile modulus of neat HDPE, untreated and alkali treated DDGS fiber composites. Different letters in the label show statistically significant difference between the different formulations at α=0.05

The DDGS fiber filled composites showed reduced tensile strength than the neat HDPE (Fig. 4.6). The tensile strength of a material is determined by the weakest part of a sample which is the interfacial region of a composite sample due to the incompatibility between the hydrophilic fiber and hydrophobic matrix. When the fiber loading increased from 25 to 50%, the tensile strength decreased as the interfacial adhesion between the fiber and the matrix further weakened. This may be due to the inadequate wetting of the fiber by the matrix, leading to crack formation in the interfacial area. Also, the higher fiber loading leads to voids in the interfacial areas in the composites [31]. The alkali treatment of fibers had no effect on the tensile strength of

composites. The tensile strength decreased by about 25% when neat HDPE was filled with 50% DDGS fiber. d b,c b,c b,c a 0 200 400 600 800 1000 1200 1400 1600 1800

Neat HDPE Untreated DDGS25 NaOH treated DDGS25 Untreated DDGS50 NaOH treated DDGS50 T en sil e S tif fn ess (M P a) Treatments

Fig. 4.6. Tensile strength of neat HDPE, untreated and alkali treated DDGS fiber composites. Different letters in the label show statistically significant difference between the different formulations at α=0.05

Flexural Properties of Composites

The inclusion of untreated DDGS fiber at 25% loading into HDPE resulted in a 23% decrease in flexural stiffness whereas all other composite treatments showed an increase in flexural stiffness (Fig. 4.7). For example, addition of treated DDGS fiber increased flexural stiffness by 24%. This increase in flexural stiffness of alkali treated DDGS25 sample was due to improved interaction between the fiber and the matrix. The modulus of a composite is

determined by the modulus of fiber and matrix, fiber content and orientation. When the fiber loading was increased to 50%, the untreated DDGS50 samples showed an increase in stiffness of 59% whereas alkali treated DDGS50 exhibited similar stiffness as of the alkali treated DDGS25 when compared to the neat HDPE.

a b b c _c 0 2 4 6 8 10 12 14 16 18 20

Neat HDPE Untreated DDGS25 NaOH treated DDGS25 Untreated DDGS50 NaOH treated DDGS50 T en sil e str en gth (M P a) Treatments

Fig. 4.7. Flexural stiffness of neat HDPE, untreated and alkali treated DDGS fiber composites. Different letters in the label show statistically significant difference between the different formulations at α=0.05

The tensile strength of DDGS composites at 25% loading showed that alkali treatment improved the tensile strength, indicating improved interfacial bonding between the fiber and the HDPE matrix (Fig. 4.8). At 50% fiber loading, the flexural strength of composites with treated and untreated DDGS fiber fillers were similar, and comparable to those of composites with 25% untreated fiber. The alkali treated DDGS25 sample had higher flexural strength than both the untreated and alkali treated DDGS50 samples by 23 and 42%, respectively.

Fig. 4.8. Flexural strength of neat HDPE, and untreated and alkali treated DDGS fiber composites. Different letters in the label show statistically significant difference between the different formulations at α=0.05 c d a,b a b 0 200 400 600 800 1000 1200

Neat HDPE Untreated DDGS25 NaOH treated DDGS25 Untreated DDGS50 NaOH treated DDGS50 F lexural S tif fn ess (M P a) Treatments a _b a _b b 0 2 4 6 8 10 12 14 16 18

Neat HDPE Untreated DDGS25 NaOH treated DDGS25 Untreated DDGS50 NaOH treated DDGS50 F lexural S tr en gth (M P a) Treatments

Impact Strength of Composites

The notched impact strengths of composite samples decreased with increase in filler loadings from 25 to 50% (Fig. 4.9). The impact strength of composites is dependent on the fiber concentration, shape, orientation, and interfacial area between the fiber and the matrix [32]. The neat HDPE sample showed the highest impact strength of 71 J/m. Addition of the fiber filler reduced the impact strength to up to one seventh. The composite samples with 25% alkali treated DDGS had higher impact strength than untreated DDSG25 indicating improvement in interfacial bonding between the treated fiber and the matrix.

Fig. 4.9. Impact strength of neat HDPE, and untreated and alkali treated DDGS fiber composites.